Staff Research Scientist: World Modeling & AI Systems

Lightfield; San Francisco (CA); Hybrid; USD 170,000 - 250,000; Full time

Benefits offered by this job

Competitive salary

Meaningful early equity

Health insurance

3 weeks PTO

11 holidays + winter break

3 months paid family leave

Wednesdays work from home

Team dinners, events, offsites, andRet

401k plan

Commuter stipend

Lunch stipend

Job summary

Lightfield is building a Business World Model through frontier research in simulation, world modeling, and self-improvement. We hire applied researchers and research scientists to lead pioneering projects, reporting to our Chief AI Officer and collaborating with engineers and GTM experts.

Researchers own projects from the initial question to measurable product impact and publication, shaping the lab's agenda and reputation as it grows. This is a full-time, hybrid role in San Francisco.

Qualifications

• 7+ years of applied research experience in ML or NLP, in industry or academia.
• PhD or MS in Computer Science, Machine Learning or a related field, or equivalent research experience.
• Track record of research that shipped or was published: papers, open-source models, datasets or benchmarks.
• Deep hands-on experience with LLMs, including evaluation, agent design and retrieval. Post-training experience (SFT, RL, preference optimization) is a plus.
• Rigorous about experimental design and able to distinguish real improvements from noise.
• Strong Python and production-code skills; comfortable in a codebase with TypeScript/Node.js/PostgreSQL.
• Able to drive ambiguous problems independently and deliver concrete results.
• Excellent written and verbal communication for technical and non-technical audiences.

Responsibilities

• Lead research projects in evaluation, benchmarks, deal prediction, simulation and self-improvement.
• Design empirical evaluations of Lightfield's AI system using LLMs as judges, rubric-based grading, offline replay and A/B tests (production and offline data).
• Build benchmarks for GTM tasks such as list building, CSV imports, mail intelligence, record updates and outbound; compare against frontier models and competitors.
• Construct simulated companies with users and data streams to evaluate the full system end to end.
• Develop techniques for better learning and adaptation, including RL and selective fine-tuning.
• Model deal probability and expose as product features; optimize cost and quality trade-offs across the AI stack.
• Publish research blog posts, open benchmarks, arXiv reports and papers; present at conferences and meetups.
• Help recruit and mentor teammates to grow a world-class research team.

Skills

Applied ML/NLP research

Python

Experiment design

Communication

Independent problem solving

LLMs experience

Education

PhD or MS in Computer Science / ML or related field

Tools

TypeScript

Node.js

PostgreSQL

Job description

Lightfield is building a Business World Model through frontier research in simulation, world modeling, and self-improvement. We hire applied researchers and research scientists to lead pioneering projects, reporting to our Chief AI Officer and collaborating with engineers and GTM experts.

Researchers own projects from the initial question to measurable product impact and publication, shaping the lab's agenda and reputation as it grows. This is a full-time, hybrid role in San Francisco.